Concept extraction and e-commerce applications

Abstract

Concept extraction is the technique of mining the most important topic of a document. In the e-commerce context, concept extraction can be used to identify what a shopping related Web page is talking about. This is practically useful in applications like search relevance and product matching. In this paper, we investigate two concept extraction methods: Automatic Concept Extractor (ACE) and Automatic Keyphrase Extraction (KEA). ACE is an unsupervised method that looks at both text and HTML tags. We upgrade ACE into Improved Concept Extractor (ICE) with significant improvements. KEA is a supervised learning system. We evaluate the methods by comparing automatically generated concepts to a gold standard. The experimental results demonstrate that ICE significantly outperforms ACE and also outperforms KEA in concept extraction. To demonstrate the practical use of concept extraction in the e-commerce context, we use ICE and KEA to showcase two e-commerce applications, i.e. product matching and topic-based opinion mining.